Researchers Information System

日本語に切り替えるswitch to english

Hayashi, Kazuki

Graduate School of Engineering, Division of Architecture and Architectural Engineering Assistant Professor

Hayashi, Kazuki
list
    Last Updated :2024/07/03

    Basic Information

    Faculty

    • Faculty of Engineering

    Professional Memberships

    • From Jun. 2022, To Present
      the Japan Society of Mechanical Engineers
    • From Jun. 2021, To Present
      International Society for Structural and Multidisciplinary Optimization (ISSMO)
    • From Jun. 2021, To Present
      International Association for Shell and Spatial Structures (IASS)
    • From Apr. 2021, To Present
      The Japan Society for Computational Engineering and Science
    • From Apr. 2015, To Present
      The Architectural Institute of Japan

    Academic Degree

    • Master of Engineering(Kyoto University)
    • Doctor of Engineering(Kyoto University)

    Research History

    • From Sep. 2023, To Aug. 2024
      École nationale des ponts et chaussées, Laboratoire Navier, Visiting researcher
    • From Apr. 2018, To Mar. 2021
      Japan Society for the Promotion of Science, Research Fellowship for Young Scientists (DC1)
    • From Apr. 2021, To Present
      Kyoto University, Graduate School of Engineering Department of Architecture and Architectual Systems Engineering, Assistant Professor

    ID,URL

    researchmap URL

    list
      Last Updated :2024/07/03

      Research

      Research Interests

      • Mathematical modelling of architectural structures
      • Structural Morphogenesis
      • Structural Optimization

      Research Areas

      • Social infrastructure (civil Engineering, architecture, disaster prevention), Building structures and materials, Structural morphogenesis
      • Social infrastructure (civil Engineering, architecture, disaster prevention), Building structures and materials, Structural optimization

      Papers

      • Architectural design education method introducing feedback by 2D mapping of design evaluations
        Hiroto Ota; Kazuki Hayashi
        AIJ Journal of Technology and Design, 20 Jun. 2024, Peer-reviewed
      • Parametric generation of optimal structures through discrete exponential functions: unveiling connections between structural optimality and discrete isothermicity
        Kazuki Hayashi; Yoshiki Jikumaru; Yohei Yokosuka; Kentaro Hayakawa; Kenji Kajiwara
        Structural and Multidisciplinary Optimization, 29 Feb. 2024, Peer-reviewed, Lead author
      • Target shape approximation of funicular surfaces by load control using RBF interpolation
        Kazuki Hayashi; Jun Yanagimuro
        International Journal of Space Structures, Jun. 2024, Peer-reviewed, Lead author
      • Multi-objective optimization of truss structure using multi-agent reinforcement learning and graph representation
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        Engineering Applications of Artificial Intelligence, Mar. 2024, Peer-reviewed
      • Deep deterministic policy gradient and graph convolutional networks for topology optimization of braced steel frames
        Chi-tathon Kupwiwat; Yuichi Iwagoe; Kazuki hayashi; Makoto Ohsaki
        Journal of Structural Engineering, B (JP), 07 Apr. 2023, Peer-reviewed
      • Deep deterministic policy gradient and graph attention network for geometry optimization of latticed shells
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        Applied Intelligence, 17 Mar. 2023, Peer-reviewed
      • Mean curvature flow for generating discrete surfaces with piecewise constant mean curvatures
        Kazuki Hayashi; Yoshiki Jikumaru; Makoto Ohsaki; Takashi Kagaya; Yohei Yokosuka
        Computer Aided Geometric Design, 20 Mar. 2023, Peer-reviewed, Lead author
      • Assembly sequence optimization of spatial trusses using graph embedding and reinforcement learning
        Kazuki Hayashi; Makoto Ohsaki; Masaya Kotera
        Journal of International Association for Shell and Spatial Structures, 01 Dec. 2022, Peer-reviewed, Invited, Lead author
      • Deep deterministic policy gradient and graph convolutional network for bracing direction optimization of grid shells
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        Frontiers in Built Environment, 23 Aug. 2022, Peer-reviewed
      • Deep reinforcement learning-based critical element identification and demolition planning of frame structures
        Shaojun Zhu; Makoto Ohsaki; Kazuki Hayashi; Shaohan Zong; Xiaonong Guo
        Frontiers of Structural and Civil Engineering, 13 Dec. 2022, Peer-reviewed
      • Graph-based reinforcement learning for discrete cross-section optimization of planar steel frames
        Kazuki Hayashi; Makoto Ohsaki
        Advanced Engineering Informatics, 14 Jan. 2022, Peer-reviewed, Lead author
      • Machine-specified ground structures for topology optimization of binary trusses using graph embedding policy network
        Shaojun Zhu; Makoto Ohsaki; Kazuki Hayashi; Xiaonong Guo
        Advances in Engineering Software, Sep. 2021, Peer-reviewed
      • Discrete Gaussian Curvature Flow for Piecewise Constant Gaussian Curvature Surface
        Kazuki Hayashi; Yoshiki Jikumaru; Makoto Ohsaki; Takashi Kagaya; Yohei Yokosuka
        Computer-Aided Design, May 2021, Peer-reviewed, Lead author
      • DESIGN REVIEW METHOD WITH “LIVE AHP” TO VISUALIZE AND SHARE JURY’S OWN DESIGN EVALUATION
        Hiroto OTA; Takuya ITO; Kazuki HAYASHI
        AIJ Journal of Technology and Design, 20 Feb. 2021, Peer-reviewed, Last author
      • Reinforcement learning and graph embedding for binary truss topology optimization under stress and displacement constraints
        Kazuki Hayashi; Makoto Ohsaki
        Frontiers in Built Environment, 30 Apr. 2020, Peer-reviewed, Lead author
      • Reinforcement learning for optimum design of a plane frame under static loads
        Kazuki Hayashi; Makoto Ohsaki
        Engineering with Computers, Jul. 2021, Peer-reviewed, Lead author
      • FDMopt: Force density method for optimal geometry and topology of trusses
        Kazuki Hayashi; Makoto Ohsaki
        Advances in Engineering Software, Jul. 2019, Peer-reviewed, Lead author
      • Force density method for simultaneous optimization of geometry and topology of trusses
        Makoto Ohsaki; Kazuki Hayashi
        Structural and Multidisciplinary Optimization, Nov. 2017, Peer-reviewed, Last author

      Misc.

      • Machine learning and optimization
        Kazuki Hayashi; Makoto Ohsaki
        Architectural Technology 2024 April issue, Mar. 2024, Lead author
      • Introduction of two-dimensional mapping for design evaluation in architectural design education - Case study of the Mitaka Satellite Campus, a second-semester design project of the second year undergraduates
        Hiroto Ota; Kazuki Hayashi
        Bulletin of Institute of Architecture, Musashino University, Mar. 2023
      • Guidelines for applying machine learning models to building design and property prediction
        Kazuki Hayashi
        "Architecture and Information―Toward the future of architectural studies", Research Council on Information System Technology Division, Architectural Institute of Japan, 06 Sep. 2022, Invited, Lead author
      • Participation report of IASS symposium 2019 in Barcelona
        Yosuke Nakaso; Tatsuya Yoshino; Kazuki Hayashi; Kenichi Kawaguchi
        Steel Structure Technology April 2020 issue, 28 Mar. 2020
      • Programming design methods
        Kazuki Hayashi
        Journal of Architecture and Building Science, Nov. 2019, Invited, Lead author

      Presentations

      • Design of bolted connections for multi-layer auxetic structures via conformal mapping
        Kazuki Hayashi; Romain Mesnil
        Annual Meeting of the Japan Society for Industrial and Applied Mathematics (JSIAM2024), Sep. 2024, the Japan Society for Industrial and Applied Mathematics (JSIAM2024)
      • Topology optimization of periodic lattice structures using machine learning considering member connectivity
        Tomoya Matsuoka; Makoto Ohsaki; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ), 30 Aug. 2024, Architectural Institute of Japan
      • Decoding human thermal sensation through artificial intelligence: a participatory experiment
        Matteo Migliari; Kazuki hayashi; Yan Ulanowski; Stéphane Laporte; Martin Hendel; Julien Despax; Loïc Chesne; Olivier Baverel
        EduBIM2024 : Données, intelligences et nature de la ville durable, Nov. 2024, EduBIM
      • Optimal design of spatial steel frames based on multi-agent reinforcement learning
        Kotaro Takenaka; Makoto Yamakawa; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 30 Aug. 2024, Architectural Institute of Japan
      • Non-developable surface structures using bilayer auxetic material and kerf bending joints
        Kazuki Hayashi; Romain Mesnil
        IASS Symposium 2024, Aug. 2024, International Association of Shell and Spatial Structures (IASS)
      • Topology optimization of periodic lattice structures for specified mechanical properties using machine learning considering member connectivity
        Tomoya Matsuoka; Makoto Ohsaki; Kazuki Hayashi
        Asian Congress of Structural and Multidisciplinary Optimization (ACSMO), 22 May 2024, Asian Society for Structural and Multidisciplinary Optimization (ASSMO)
      • Evaluating generalization performance of boundary-informed cGANs for continuum topology optimization
        On Takahashi; Kazuki Hayashi; Makoto Ohsaki
        Asian Congress of Structural and Multidisciplinary Optimization (ACSMO), 22 May 2024, Asian Society for Structural and Multidisciplinary Optimization (ASSMO)
      • At the intersection of structural engineering, discrete differential geometry, and machine learning
        Kazuki Hayashi
        PhD seminar at TU Munich, 21 Mar. 2024, Technical University of Munich, Invited
      • Topology design of lattice structures with specified deformations using machine learning
        Tomoya Matsuoka; Makoto Ohsaki; Kazuki Hayashi
        The 46th Symposium on Computer Technology of Information, Systems and Applications, 07 Dec. 2023, Architectural Institute of Japan
      • Graph is a key to extracting knowledge for structural optimization and architectural design
        Kazuki Hayashi
        COAST seminar, 05 Dec. 2023, Navier Laboratory, École nationale des ponts et chaussées, Invited
      • Topology optimization of continuum structure using conditional generative adversarial network with boundary conditions as input
        On Takahashi; Kazuki Hayashi; Makoto Ohsaki
        the 18th colloquium analysis and generation of structural shapes and systems, 17 Nov. 2023, Architectural Institute of Japan
      • Optimal design of steel frames based on policy network reinforcement learning
        Kotaro Takenaka; Makoto Yamakawa; Kazuki Hayashi
        the 33rd Design and Systems Conference (D&S2022), 20 Sep. 2023, The Japan Society of Mechanical Engineers (JSME)
      • Extracting meaningful features from data with irregular connectivity for structural optimization
        Kazuki Hayashi
        Scientific Computing Seminar, 09 Nov. 2023, University of Kaiserslautern-Landau, Invited
      • Shape optimization of shell structures consisting of discrete surfaces with piecewise constant Gaussian curvature
        Kenta Fukui; Makoto Ohsaki; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ), 13 Sep. 2023, Architectural Institute of Japan
      • A multi-agent reinforcement learning framework for bi-objective optimization
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        the 33th Design and Systems Conference (D&S2023), 21 Sep. 2023, The Japan Society of Mechanical Engineers
      • Multi-objective optimization of 10-bar truss using multi-agent reinforcement learning
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 14 Sep. 2023, Architectural Institute of Japan
      • Topology optimization of continuum structures using conditional generative adversarial networks
        On Takahashi; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 15 Sep. 2023, Architectural Institute of Japan
      • Optimization of lattice structures by machine learning models using feature selection
        Tomoya Matsuoka; Makoto Ohsaki; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ), 15 Sep. 2023, Architectural Institute of Japan
      • Mean curvature flow for generating triangular meshes with piecewise constant mean curvatures
        Kazuki Hayashi; Yoshiki Jikumaru; Makoto Ohsaki; Takashi Kagaya; Yohei Yokosuka
        the XI edition of the conference on Textile Composites and Inflatable Structures (STRUCTURAL MEMBRANES 2023), 03 Oct. 2023, Carlos Lázaro, Riccardo Rossi and Roland Wüchner
      • Force density method for simultaneous optimization of geometry and topology of trusses with uniform cross- section
        Saku Aoyagi; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 14 Sep. 2023, Architectural Institute of Japan
      • Variational principle for generating discrete surfaces with piecewise constant Gaussian curvatures
        Kazuki Hayashi; Yoshiki Jikumaru; Makoto Ohsaki; Takashi Kagaya; Yohei Yokosuka
        the 10h International Congress on Industrial and Applied Mathematics (ICIAM 2023), 25 Aug. 2023, the International Council for Industrial and Applied Mathematics (ICIAM)
      • Sizing optimization of free-form lattice shells using deep deterministic policy gradient and graph convolutional networks
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        IASS Symposium 2023, 11 Jul. 2023, International Association of Shell and Spatial Structures (IASS)
      • Isogeometric deep learning framework to predict the structural performance of free-form surfaces
        Kazuki Hayashi; Makoto Ohsaki
        IASS Symposium 2023, 13 Jul. 2023, International Association of Shell and Spatial Structures (IASS)
      • Study on shape generation and structural properties of discrete CGC surface generated by Gaussian curvature flow considering movement of internal boundary
        Kenta Fukui; Makoto Ohsaki; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, 24 Jun. 2023, Architectural Institute of Japan
      • Target shape approximation of anti-funicular shells by load control using RBF interpolation
        Kazuki Hayashi; Jun Yanagimuro
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, 24 Jun. 2023, Architectural Institute of Japan
      • Reinforcement learning and graph representations for optimization of plane steel building frames
        Makoto Ohsaki; Kazuki Hayashi; Chi-tathon Kupwiwat
        the 15th World Congress on Structural and Multidisciplinary Optimisation (WCSMO2023), 09 Jun. 2023, the international Society for Structural and Multidisciplinary Optimisation (ISSMO)
      • A truss structure with mechanical optimality, artisiticity and integrability
        Kenji Kajiwara; Yohei Yokosuka; Kazuki Hayashi; Yoshiki Jikumaru; Kentaro Hayakawa
        Symmetries and Integrability of Difference Equations 14.2, 21 Jun. 2023, Symmetries and Integrability of Difference Equations
      • On the Michell truss-like structure and discrete log-aesthetic curves based on integrable geometry
        Yoshiki Jukumaru; Kentaro Hayakawa; Kazuki Hayashi; Kenji Kajiwara; Yohei Yokosuka
        Annual meeting of The Mathematical Society of Japan, 17 Mar. 2023, The Mathematical Society of Japan
      • Generating kinematically-stable 2d trusses for topology optimization by reinforcement learning and graph embedding
        Shaojun Zhu; Makoto Ohsaki; Kazuki Hayashi
        IASS 2022 Symposium affiliated with APCS 2022 conference, 20 Sep. 2022, International Association for Shell and Spatial Structures (IASS)
      • Simultaneous generation of shape and stress distribution by discrete membrane O-surface
        Yohei Yokosuka; Yoshiki Jikumaru; Kazuki Hayashi; Kentaro Hayakawa; Takanori Yagi; Shono Suzuki; Yusuke Sakai; Keisuke Mizutani
        Annual Convention of the Japan Society for Industrial and Applied Mathematics, 10 Sep. 2022, the Japan Society for Industrial and Applied Mathematics
      • Structural performance of triangular latticed shells with regularized panels for Bézier design surfaces
        Kazuki Hayashi; Makoto Ohsaki
        12th Asian Pacific Conference on Shell and Spatial Structures (APCS2018), 29 Oct. 2018, The School of Civil Engineering, Universiti Sains Malaysia
      • FDMOPT: A new tool for simultaneous optimization of geometry
        Makoto Ohsaki; Kazuki Hayashi; Caitlin Mueller
        IASS annual symposium 2018, 19 Jul. 2018, International Association for Shell and Spatial Structures (IASS)
      • Regularization of triangular latticed shell panels for Bézier surfaces
        Kazuki Hayashi; Makoto Ohsaki
        IASS annual symposium 2018, 17 Jul. 2018, International Association for Shell and Spatial Structures (IASS)
      • Force density method for simultaneous optimization of geometry and topology of spatial trusses
        Kazuki Hayashi; Makoto Ohsaki; Caitlin Mueller
        IASS annual symposium 2017, Sep. 2017, International Association for Shell and Spatial Structures (IASS)
      • On the Michell truss-like structures recovered by discrete holomorphic functions
        Kazuki Hayashi; Kentaro Hayakawa; Yoshiki Jikumaru; Kenji Kajiwara; Yohei Yokosuka
        Annual Convention of the Japan Society for Industrial and Applied Mathematics, 10 Sep. 2022, the Japan Society for Industrial and Applied Mathematics
      • Simultaneous optimization of discrete cross-sections and nodal coordinates for spatial trusses combining time-window particle swarm optimization and covariance matrix adaptation evolution strategy
        Kazuki Hayashi; Makoto Ohsaki
        The 14th Optimization Symposium 2022 (OPTIS2022), 12 Nov. 2022, The Japan Society of Mechanical Engineers
      • Topology optimization of planar trusses combining structural analysis by classical computing and cross-section change by quantum annealing
        Kazuki Hayashi; Makoto Ohsaki
        The 45th Symposium on Computer Technology of Information, Systems and Applications, 01 Dec. 2022, Architectural Institute of Japan
      • Gaussian curvature flow for constant Gaussian curvature surface with triangular mesh
        Makoto Ohsaki; Kazuki Hayashi; Yoshiki Jikumaru; Takashi Kagaya; Yohei Yokosuka
        IASS 2022 Symposium affiliated with APCS 2022 conference, 21 Sep. 2022, International Association of Shell and Spatial Structures (IASS)
      • Deep deterministic policy gradient and graph convolutional network for geometry and topology optimization of braced latticed shells
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, 26 Jun. 2022, Architectural Institute of Japan (AIJ)
      • Low-dimensional mapping of design evaluation data by analytic hierarchy process and machine learning
        Kazuki Hayashi; Hiroto Ota
        the 32th Design and Systems Conference (D&S2022), 22 Sep. 2022, The Japan Society of Mechanical Engineers (JSME)
      • Topology optimization of braced latticed shells using deep deterministic policy gradient and graph convolutional network
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        Asian Congress of Structural and Multidisciplinary Optimization (ACSMO), 25 May 2022, Asian Society for Structural and Multidisciplinary Optimization (ASSMO)
      • Generation of discrete surfaces with piecewise constant mean curvature by mean curvature flow considering boundary terms
        Kazuki Hayashi; Yoshiki Jikumaru; Makoto Ohsaki; Takashi Kagaya; Yohei Yokosuka
        the 18th union presentation of research groups, 08 Mar. 2022, the Japan Society for Industrial and Applied Mathematics
      • Reinforcement learning for construction process of truss by stability assessment and graph embedding
        Masaya Kotera; Kazuki Hayashi; Makoto Ohsaki
        The 44th Symposium on Computer Technology of Information, Systems and Applications, 09 Dec. 2021, Architectural Institute of Japan
      • Topology optimization of planar trusses by reinforcement learning agents trained for the sequential member addition and removal process
        Kazuki Hayashi; Makoto Ohsaki
        The 44th Symposium on Computer Technology of Information, Systems and Applications, 09 Dec. 2021, Architectural Institute of Japan
      • A method for simultaneous generation of optimal shapes of planar trusses with different aspect ratios by multi-objective optimization
        Saku Aoyagi; Kazuki Hayashi; Makoto Ohsaki
        the 16th colloquium analysis and generation of structural shapes and systems, 29 Oct. 2021
      • Reinforcement learning and graph embedding for generating stable construction process of trusses
        Masaya Kotera; Makoto Ohsaki; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ), 09 Sep. 2021, Architecture Institute of Japan
      • Force density method for simultaneous optimization of geometry and topology of curved surface trusses
        Saku Aoyagi; Makoto Ohsaki; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ), 09 Sep. 2021, Architecture Institute of Japan
      • Optimal topology prediction of planar trusses using graph embedding and supervised learning
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 07 Sep. 2021, Architecture Institute of Japan
      • Graph embedding and machine learning for topology optimization of planar trusses
        Kazuma Sakaguchi; Makoto Ohsaki; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ), 07 Sep. 2021, Architecture Institute of Japan
      • Reinforcement learning for optimal topology design of 3D trusses
        Kazuki Hayashi; Makoto Ohsaki
        IASS Annual Symposium 2020/21, 27 Aug. 2021, International Association of Shell and Spatial Structures (IASS)
      • Reinforcement learning for topology optimization of 3D trusses using graph embedding and block matrix
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, 26 Jun. 2021, Architectural Institute of Japan
      • Machine learning and graph embedding for truss topology optimization
        Makoto Ohsaki; Kazuki Hayashi; Kazuma Sakaguchi
        14th World Congress of Structural and Multidisciplinary Optimization (WCSMO-14), 16 Jun. 2021, International Society for Structural and Multidisciplinary Optimization (ISSMO)
      • Cross-section optimization of steel frames using graph-based reinforcement learning
        Kazuki Hayashi; Makoto Ohsaki
        14th World Congress of Structural and Multidisciplinary Optimization (WCSMO-14), 15 Jun. 2021, International Society for Structural and Multidisciplinary Optimization (ISSMO)
      • Gaussian curvature flow for architectural morphogenesis of discrete surfaces with piece-wisely constant Gaussian curvature
        Kazuki Hayashi; Yoshiki Jikumaru; Makoto Ohsaki; Takashi Kagaya; Yohei Yokosuka
        the 17th union presentation of research groups, 05 Mar. 2021, the Japan Society for Industrial and Applied Mathematics
      • Minimum-volume design of steel frames using reinforcement learning
        Kazuki Hayashi; Makoto Ohsaki
        ECCOMAS Congress 2020 and 14th WCCM, 14 Jan. 2021, International Association for Computational Mechanics (IACM) and European Community on Computational Methods in Applied Sciences (ECCOMAS)
      • Development of an agent for discrete cross-section design of planar steel frames using graph embedding and reinforcement learning
        Kazuki Hayashi; Makoto Ohsaki
        The 43rd Symposium on Computer Technology of Information, Systems and Applications, 10 Dec. 2020
      • Reinforcement learning and graph embedding for sequential optimal design of plane trusses and frames
        Kazuki Hayashi; Makoto Ohsaki
        Asian Congress of Structural and Multidisciplinary Optimization (ACSMO), 25 Nov. 2020, Asian Society of Structural and Multidisciplinary Optimization (ASSMO) and Korean Society for Design Optimization (KSDO)
      • Graph embedding and reinforcement learning for topology optimization of planar trusses with stress constraints
        Kazuki hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, Jun. 2020, Architectural Institute of Japan
      • Study on communication environment to visualize and share jury’s own ”design evaluation” - Design review system with AHP application
        Hiroto Ota; Takuya Ito; Kazuki Hayashi
        Annual Convention of Architectural Institute of Japan (AIJ), Sep. 2020, Architectural Institute of Japan
      • Reinforcement learning and graph embedding for cross-sectional design of planar steel frames
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), Sep. 2020, Architectural Institute of Japan
      • Development of an application to visualize and share user’s own design review expressed as AHP - A report of experiment in the review-event “Diploma x Kyoto”
        Hiroto Ota; Kazuki Hayashi
        Design Symposium, 17 Nov. 2019
      • Deep-Q network for truss topology optimization with stress constraints
        Kazuki Hayashi; Makoto Ohsaki
        IASS symposium, 09 Oct. 2019, International Association of Shell and Spatial Structures (IASS)
      • Shape optimization of frame structures using dynamic programming
        Kazuki Hayashi; Makoto Ohsaki
        The 41st Symposium on Computer Technology of Information, Systems and Applications, 06 Dec. 2018, Architectural Institute of Japan
      • Force density method for simultaneous optimization of geometry and topology for latticed shells with free-form design surface
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 05 Sep. 2018
      • Regularization of edge length of triangular panels for latticed shells with free-form surfaces
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, 24 Jun. 2018, Architectural Institute of Japan
      • Simultaneous topology and shape optimization by force density method - Analysis of constraint method for member length and nodal position
        Kazuki Hayashi; Makoto Ohsaki
        The 40th Symposium on Computer Technology of Information, Systems and Applications, 14 Dec. 2017
      • Force density method for simultaneous optimization of geometry and topology of trusses
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 03 Sep. 2017, Architectural Institute of Japan
      • Simultaneous optimization of topology and geometry of planar truss using force density as design variable
        Kazuki Hayashi; Makoto Ohsaki
        The 11th colloquium analysis and generation of structural shapes and systems, 28 Oct. 2016, Architectural Institute of Japan
      • A study on critique of architectural and urban design by the method of using AHP
        Kazuki Hayashi; Hiroto Ota
        Annual Convention of Architectural Institute of Japan (AIJ), 06 Sep. 2015, Architectural Institute of Japan
      • Assembly sequence optimization of spatial trusses using graph embedding and reinforcement learning
        Kazuki Hayashi; Makoto Ohsaki; Masaya Kotera
        IASS 2022 Symposium affiliated with APCS 2022 conference, 22 Sep. 2022, International Association for Shell and Spatial Structures (IASS), Invited
      • Design of brace location and cross-section of steel frame by reinforcement learning using grid features
        Yuichi Iwagoe; Makoto Ohsaki; Kazuki Hayashi; Kupwiwat Chi-tathon
        Annual Convention of Architectural Institute of Japan (AIJ), 07 Sep. 2022, Architectural Institute of Japan
      • Geometry optimization of lattice shells using GAT-DDPG with Bézier surface
        Chi-tathon Kupwiwat; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 08 Sep. 2022, Architectural Institute of Japan
      • Generation method of slit patterns using Bézier curves and structural analysis models with beam elements for kerf bending
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 08 Sep. 2022, Architectural Institute of Japan
      • Graph and machine learning-based approach to prediction of ultimate load of latticed shells considering geometric nonlinearity
        Kazuki Hayashi; Makoto Ohsaki
        15th World Congress on Computational Mechanics (WCCM-XV) 8th Asian Pacific Congress on Computational Mechanics (APCOM-VIII), 03 Aug. 2022, The International Association for Computational Mechanics (IACM) and The Japan Society for Computational Engineering and Science (JSCES) in cooperation with the Asian-Pacific Association for Computational Mechanics (APACM) and Japan Association for Computational Mechanics (JACM)
      • Generating optimal shape of plane truss with various aspect ratios by multi-objective optimization
        Saku Aoyagi; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, 26 Jun. 2022, Architectural Institute of Japan
      • Prediction of the reduction factors for elastic buckling loads of single-layer latticed shells using graph embedding and machine learning
        Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ) Kinki Branch, 26 Jun. 2022, Architectural Institute of Japan
      • Machine learning model using graph embedding for extracting features of skeletal structures
        Kazuki Hayashi; Makoto Ohsaki
        The 66th National Congress of Theoretical and Applied Mechanics, 24 Jun. 2022, The Japan Federation of Engineering Societies (JFES) Theoretical Applied Mechanics Consortium
      • Force density method for simultaneous optimization of geometry and topology of trusses of uniform cross-sections on free-form design surface
        Saku Aoyagi; Kazuki Hayashi; Makoto Ohsaki
        Asian Congress of Structural and Multidisciplinary Optimization (ACSMO), 24 May 2022, Asian Society for Structural and Multidisciplinary Optimization (ASSMO)
      • Knowledge extraction of discrete cross-section optimization of planar steel frames using graph-based reinforcement learning
        Kazuki Hayashi; Makoto Ohsaki
        Asian Congress of Structural and Multidisciplinary Optimization (ACSMO), 24 May 2022, Asian Society for Structural and Multidisciplinary Optimization (ASSMO)
      • Predictive model for external force work of trusses using graph embedding and machine learning
        Keiya Nakazato; Kazuki Hayashi; Makoto Ohsaki
        Annual Convention of Architectural Institute of Japan (AIJ), 07 Sep. 2022, Architectural Institute of Japan
      • Differential geometric approaches to shell membrane theory and their applications to architectural surface design
        Kentaro Hayakawa; Yoshiki Jikumaru; Yohei Yokosuka; Takashi Kagaya; Kazuki Hayashi; Yusuke Sakai, Joint editor
        Math-for-industry Education and Research Hub, Kyushu University, 28 Jul. 2021

      Works

      • RBF Load
        Kazuki Hayashi; Jun Yanagimuro
        From 31 Jul. 2023, To Present
      • LIVE AHP
        Kazuki Hayashi; Hiroto Ota
        From 29 Apr. 2020, To Present
      • Assembly Sequence Predictor
        Kazuki Hayashi; Makoto Ohsaki
        From 27 Apr. 2022, To Present
      • FDMopt
        Kazuki Hayashi; Makoto Ohsaki
        From 15 Sep. 2021, To Present

      External funds: Kakenhi

      • Development of a combined method of graph embedding and machine learning for optimal design of skeletal structures
        Grant-in-Aid for Research Activity Start-up
        0304:Architecture, building engineering, and related fields
        Kyoto University
        Kazuki Hayashi
        From 30 Aug. 2021, To 31 Mar. 2023, Project Closed
        構造最適化;機械学習;グラフ埋め込み;幾何学的深層学習;離散構造物;強化学習;教師あり学習;建築構造最適化;鋼構造平面骨組;断面最適化;非線形問題
      • Analysis and design of building frames using machine learning considering uncertainty of parameters
        Grant-in-Aid for Scientific Research (C)
        Basic Section 23010:Building structures and materials-related
        Kyoto University
        大崎 純
        From 01 Apr. 2023, To 31 Mar. 2026, Granted
        建築骨組;機械学習;構造設計;不確定性;構造最適化
      • Construction of an AI-supported system for the design and assembly sequence planning of multi-member interlocking joints
        Grant-in-Aid for Early-Career Scientists
        Basic Section 23010:Building structures and materials-related
        Kyoto University
        林 和希
        From 01 Apr. 2024, To 31 Mar. 2028, Granted
        篏合;離散構造;グラフ理論;機械学習
      list
        Last Updated :2024/07/03

        Education

        Teaching subject(s)

        • From 01 Apr. 2024, To 31 Mar. 2025
          Mechanics of Building Structures II
          4012, Fall, Faculty of Engineering, 2
        • From 01 Apr. 2023, To 31 Mar. 2024
          Mechanics of Building Structures II
          4012, Fall, Faculty of Engineering, 2
        • From 01 Apr. 2023, To 31 Mar. 2024
          Mechanics of Building StructuresI
          4011, Spring, Faculty of Engineering, 2
        • From 01 Apr. 2023, To 31 Mar. 2024
          Practice of Basic Informatics (Faculty of Engineering)
          T006, Spring, Institute for Liberal Arts and Sciences, 2
        • From 01 Apr. 2023, To 31 Mar. 2024
          Mechanics of Building StructuresIII
          4022, Spring, Faculty of Engineering, 4
        • From 01 Apr. 2022, To 31 Mar. 2023
          Mechanics of Building StructuresIII
          4022, Spring, Faculty of Engineering, 4
        • From 01 Apr. 2022, To 31 Mar. 2023
          Mechanics of Building StructuresI
          4011, Spring, Faculty of Engineering, 2
        • From 01 Apr. 2022, To 31 Mar. 2023
          Practice of Basic Informatics (Faculty of Engineering)
          T006, Spring, Institute for Liberal Arts and Sciences, 2
        • From Apr. 2021, To Mar. 2022
          Mechanics of Building StructuresIII
          Spring, 工学部
        • From Apr. 2021, To Mar. 2022
          Mechanics of Building StructuresI
          Spring, 工学部
        list
          Last Updated :2024/07/03

          Academic, Social Contribution

          Committee Memberships

          • From Nov. 2023, To Jun. 2025
            Local Scientific Committee, The 14th Asian Pacific Conference on Shell and Spatial Structures(APCS2025)
          • From Apr. 2023, To Present
            member, The Japan Institute of Architects - subcommittee on structural optimization and integrated design
          • From Apr. 2022, To Present
            co-chair, IASS WG13 Study Group Next Generation Parametric Design
          • From Aug. 2022, To Jan. 2024
            Member of the Public Relations Committee of the annual meeting (2023) of Architectural Institute of Japan, Architectural Institute of Japan
          • From Apr. 2021, To Mar. 2023
            member, The Japan Institute of Architects - subcommittee on structural optimization and collaborative creation

          Academic Contribution

          • WG13 session "Numerical Methods for geometry, form-finding and optimization of lightweight structures", IASS 2024
            Panel moderator, Peer review
            Philippe Block, Jacqueline Pauli, Catherine De Wolf and Walter Kaufmann, Zurich, Switzerland, From 26 Aug. 2024, To 30 Aug. 2024
          • IS10 - Parametric Design of Tensile and Membrane Structures, STRUCTURAL MEMBRANES 2023
            Panel moderator, Peer review
            Carlos Lázaro, Riccardo Rossi, and Roland Wüchner, Valencia, Spain, From 02 Oct. 2023, To 04 Oct. 2023

          Social Contribution

          • Conformal Geometry Processing for Architectural/Structural Design
            Lecturer, Performer
            Navier Laboratory, École nationale des ponts et chaussées, MANGT class, From 15 May 2024, To 16 May 2024
          • Application of discrete differential geometry and machine learning for structural engineering problems
            Lecturer
            Ellioth, Ellioth University (In-office talk at Ellioth), Montreuil, France, From 17 Apr. 2024, To 17 Apr. 2024
          • Introduction to graph-based machine learning for solving structural engineering problems
            Lecturer, Performer
            Navier Laboratory, École nationale des ponts et chaussées, MSA workshop, From 05 Mar. 2024, To 05 Mar. 2024
          • Machine learning model for data with irregular connectivity
            Lecturer
            Kazuki Hayashi, the 4th Insight Data Science Conference, Online event organized by Insight Co., Ltd., From 09 Aug. 2022, To 09 Aug. 2022
          • Designing cross-sections using a machine learning model
            Lecturer, Performer
            Architectural Institute of Japan (AIJ), Symposium "Collaborative use of structural optimization", Tokyo, Japan, From 12 May 2023, To 12 May 2023
          • Demonstration of machine learning for architectural design
            Lecturer, Performer
            Subcommittee on Design Science Mathematical Intelligence, Architectural Institute of Japan, Design Science Workshop - Introduction to Design and Programming (Generation, Optimization, Machine Learning), Tokyo Institute of Technology, From 01 Oct. 2022, To 02 Oct. 2022

          ページ上部へ戻る